A Systematic Study on a Customer’s Next-Items Recommendation Techniques
Abstract
:1. Introduction
2. Related Work
3. Methodology
3.1. Planning the Review
- Defining the need for the review
- Formulating the research questions
- Developing a review protocol
3.1.1. Defining the Need for the Review
3.1.2. Formulating the Research Questions
- Population—Shopping baskets of customers;
- Intervention—Customers’ next-items recommender systems;
- Comparison—None;
- Outcome—Comparative study of next-items recommender systems;
- Context—Works related to next-items recommender systems in various domains;
- RQ1: How can a next-items recommender system support businesses and individuals in their tasks?
- RQ2: Which conventional machine learning techniques have been proposed for next-items recommender systems?
- RQ3: Which contemporary deep learning techniques have been employed for next-items recommender systems?
- RQ4: Which datasets have been used by researchers to implement their proposed solutions?
- RQ5: In addition to the general evaluation measures used for prediction problems, which special measures have been proposed by researchers for evaluating the next-items recommender systems?
- RQ6: What are some challenges and open problems in the next-items recommender systems?
3.1.3. Developing a Review Protocol
- IEEE Xplore (https://ieeexplore.ieee.org/, accessed on 7 June 2022);
- Science Direct (https://www.sciencedirect.com/, accessed on 7 June 2022);
- Springer Link (https://link.springer.com/, accessed on 7 June 2022);
- ACM Digital Library (https://dl.acm.org/, accessed on 7 June 2022);
- Web of Science (https://clarivate.com/, accessed on 7 June 2022);
- Scopus (http://scopus.com/, accessed on 7 June 2022).
- The terms “predict,” “prediction,” “recommender,” and “recommendation systems” for predicting the future shopping basket of a customer;
- The terms “grocery,” “e-commerce,” and “shopping” for recommendations in the domain of shopping;
- The terms “next-items” and “next basket” for predicting the next-items list of items for customers;
- The terms “collaborative filtering”, “deep learning”, “kNN”, “neural networks”, and “CNN” for specialized recommendation systems;
- The term “machine learning” for survey papers or general solutions to the problem.
- (recommender OR recommendation) AND ((e-commerce OR grocery OR shopping OR (next AND (item OR basket))) OR (predicting AND customer AND shopping);
- (collaborative filtering) AND (Predict*) AND (e-commerce OR grocery OR shopping OR (next AND (item OR basket)));
- (machine learning) AND (prediction AND customer AND shopping) AND (e-commerce OR (next AND (item OR basket)));
- (deep learning) AND (predict* AND customer AND shop*) AND (e-commerce OR grocery OR shopping OR (next AND (item OR basket))).
- Peer-reviewed papers that have been published;
- Papers published between 2017 and 2022 to ensure the inclusion of recent studies;
- Papers published in journals and conferences only;
- Papers written in the English language;
- Papers related to customers’ next-items recommender systems;
- Papers related to one or more research questions of this study;
- Studies presenting recommender systems for customers’ next-items recommendations using any conventional machine learning approach;
- Studies presenting recommender systems for customers’ next-items recommendations using any deep learning approach.
- Papers not related to next-items recommender systems;
- Papers published in any language other than English;
- Papers published prior to 2017;
- Papers published in journals or conferences not following a peer-review process;
- Papers unrelated to any one of the research questions of this study;
- Any types of documents other than research articles such as thesis, white papers, reports, commentaries, and editorials;
- Repeated papers found in more than one source.
- The study has well-defined objectives.
- The study reports the model and findings consistently and coherently.
- The research methods and process are detailed clearly.
- The study applies the proposed model/algorithm to a dataset.
- The evaluation metrics are clearly described and measured.
- The findings of the study are credible.
- The findings of the study are important.
- The study compares its findings with the most suitable and most recent alternatives.
- The study provides sufficient information to replicate its findings.
3.2. Conducting the Review
3.3. Reporting the Review
4. Conventional Techniques for Customer’s Next-Items Recommendation
4.1. Markov Chains (MC)
4.2. Collaborative Filtering
4.2.1. User-Based Collaborative Filtering (UBCF)
4.2.2. Item-Based Collaborative Filtering (IBCF)
4.2.3. Pros and Cons of CF Approaches
4.3. kNN
5. Deep Learning Techniques for Customer’s Next-Items Recommendation
- (i)
- Deep learning (DL) enables modeling of nonlinear interactions found in data using nonlinear activations (e.g., ReLU and Sigmoid). Unlike conventional methods that are linear in nature, nonlinear modeling of interactions allows capturing of more complex user–item interaction patterns. Consequently, users’ preferences can be more precisely reflected.
- (ii)
- DL techniques can learn the descriptive information about items and users efficiently and thus enhance the recommender system’s understanding of items and users.
- (iii)
- DL techniques have been shown to be a perfect fit for sequential modeling tasks (e.g., natural language processing) and thus work well for the temporal dynamics associated with user and item behavior.
- (iv)
- DL is highly flexible and allows combining different neural structures to construct more powerful hybrid models. This capability can be exploited to develop hybrid recommendation models that can capture and process varying characteristics simultaneously.
5.1. CNN
5.2. RNN
5.3. Graph Neural Networks
5.4. Other Deep Networks (DNN)
5.5. Hybrid Networks (Attention+)
6. Datasets
7. Evaluation Metrics
7.1. Rating Prediction Measures
7.2. Classification Accuracy Metrics
7.3. Ranking Metrics
7.4. Other Metrics
8. Challenges and Future Directions
9. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Andry, J.F.; Riwanto, R.E.; Wijaya, R.L.; Prawoto, A.A.; Prayogo, T. Development Point of Sales Using SCRUM Framework. J. Syst. Integr. 2019, 10, 1804–2724. [Google Scholar]
- Koch, J.; Frommeyer, B.; Schewe, G. Online Shopping Motives during the COVID-19 Pandemic—Lessons from the Crisis. Sustainability 2020, 12, 10247. [Google Scholar] [CrossRef]
- Gorgoglione, M.; Panniello, U.; Tuzhilin, A. Recommendation strategies in personalization applications. Inf. Manag. 2019, 56, 103143. [Google Scholar] [CrossRef]
- Akter, T. Online shopping behavior: An in-depth study on motivating and restraining factors. Glob. J. Manag. Bus. Res. 2018, 18, 19. [Google Scholar]
- Anwar, M.; Khan, S.Z.; Shah, S.Z.A. Big Data Capabilities and Firm’s Performance: A Mediating Role of Competitive Advantage. J. Inf. Knowl. Manag. 2018, 17, 1850045. [Google Scholar] [CrossRef]
- Rosa, R.L.; Schwartz, G.M.; Ruggiero, W.V.; Rodriguez, D.Z. A Knowledge-Based Recommendation System That Includes Sentiment Analysis and Deep Learning. IEEE Trans. Ind. Inform. 2018, 15, 2124–2135. [Google Scholar] [CrossRef]
- Yochum, P.; Chang, L.; Gu, T.; Zhu, M. Linked Open Data in Location-Based Recommendation System on Tourism Domain: A Survey. IEEE Access 2020, 8, 16409–16439. [Google Scholar] [CrossRef]
- Yu, F.; Liu, Q.; Wu, S.; Wang, L.; Tan, T. A dynamic recurrent model for next basket recommendation. In Proceedings of the 39th International ACM SIGIR Conference on Research and Development in Information Retriev-al, Pisa, Italy, 17–21 July 2016; pp. 729–732. [Google Scholar]
- Lee, A.J.; Wang, C.-S.; Weng, W.-Y.; Chen, Y.-A.; Wu, H.-W. An efficient algorithm for mining closed inter-transaction itemsets. Data Knowl. Eng. 2008, 66, 68–91. [Google Scholar] [CrossRef]
- Peker, S.; Kocyigit, A.; Eren, P.E. A hybrid approach for predicting customers’ individual purchase be-havior. Kybernetes 2017, 46. [Google Scholar] [CrossRef]
- Kitchenham, B.; Charters, S. Guidelines for Performing Systematic Literature Reviews in Software Engineering; EBSE Technical Report EBSE-2007-01; Keele University: Keele, UK, 2007. [Google Scholar]
- Alyari, F.; Navimipour, N.J. Recommender systems. Kybernetes 2018, 47, 985–1017. [Google Scholar] [CrossRef]
- Portugal, I.; Alencar, P.; Cowan, D. The use of machine learning algorithms in recommender systems: A systematic review. Expert Syst. Appl. 2018, 97, 205–227. [Google Scholar] [CrossRef]
- Monti, D.; Rizzo, G.; Morisio, M. A systematic literature review of multicriteria recommender systems. Artif. Intell. Rev. 2020, 54, 427–468. [Google Scholar] [CrossRef]
- Villegas, N.M.; Sánchez, C.; Díaz-Cely, J.; Tamura, G. Characterizing context-aware recommender systems: A systematic literature review. Knowl. -Based Syst. 2018, 140, 173–200. [Google Scholar] [CrossRef]
- Murciego, L.; Jiménez-Bravo, D.; Román, A.V.; Santana, J.D.P.; Moreno-García, M. Context-Aware Recommender Systems in the Music Domain: A Systematic Literature Review. Electronics 2021, 10, 1555. [Google Scholar] [CrossRef]
- Jesse, M.; Jannach, D. Digital nudging with recommender systems: Survey and future directions. Comput. Hum. Behav. Rep. 2021, 3, 100052. [Google Scholar] [CrossRef]
- Da’U, A.; Salim, N. Recommendation system based on deep learning methods: A systematic review and new directions. Artif. Intell. Rev. 2019, 53, 2709–2748. [Google Scholar] [CrossRef]
- Khan, H.R.; Lim, C.; Ahmed, M.; Tan, K.; Bin Mokhtar, M. Systematic Review of Contextual Suggestion and Recommendation Systems for Sustainable e-Tourism. Sustainability 2021, 13, 8141. [Google Scholar] [CrossRef]
- Hamid, R.A.; Albahri, A.; Alwan, J.K.; Al-Qaysi, Z.; Zaidan, A.; Alnoor, A.; Alamoodi, A.; Zaidan, B. How smart is e-tourism? A systematic review of smart tourism recommendation system applying data management. Comput. Sci. Rev. 2020, 39, 100337. [Google Scholar] [CrossRef]
- Mohammadi, V.; Rahmani, A.M.; Darwesh, A.M.; Sahafi, A. Trust-based recommendation systems in Internet of Things: A systematic literature review. Hum.-Cent. Comput. Inf. Sci. 2019, 9, 21. [Google Scholar] [CrossRef]
- Rahayu, N.W.; Ferdiana, R.; Kusumawardani, S.S. A systematic review of ontology use in E-Learning recommender system. Comput. Educ. Artif. Intell. 2022, 3, 100047. [Google Scholar] [CrossRef]
- Wohlin, C.; Runeson, P.; Höst, M.; Ohlsson, M.C.; Regnell, B.; Wesslén, A. Experimentation in Software Engineering; Springer: Berlin/Heidelberg, Germany, 2012. [Google Scholar] [CrossRef]
- Petticrew, M.; Roberts, H. Systematic Reviews in the Social Sciences: A Practical Guide; John Wiley & Sons: Hoboken, NJ, USA, 2008. [Google Scholar]
- Page, M.J.; McKenzie, J.E.; Bossuyt, P.M.; Boutron, I.; Hoffmann, T.C.; Mulrow, C.D.; Shamseer, L.; Tetzlaff, J.M.; Akl, E.A.; Brennan, S.E.; et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. Int. J. Surg. 2021, 88, 105906. [Google Scholar] [CrossRef] [PubMed]
- Zhao, C.; You, J.; Wen, X.; Li, X. Deep Bi-LSTM Networks for Sequential Recommendation. Entropy 2020, 22, 870. [Google Scholar] [CrossRef] [PubMed]
- He, R.; Kang, W.-C.; McAuley, J. Translation-based Recommendation: A Scalable Method for Modeling Sequential Behavior. In Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence Best Sister Conferences, Stockholm, Sweden, 13–19 July 2018; pp. 5264–5268. [Google Scholar] [CrossRef]
- He, R.; Kang, W.-C.; McAuley, J. Translation-based Recommendation. In Proceedings of the Eleventh ACM Conference on Recommender Systems, Como, Italy, 27–31 August 2017; pp. 161–169. [Google Scholar] [CrossRef]
- Huang, T.; Zhang, D.; Bi, L. Neural embedding collaborative filtering for recommender systems. Neural Comput. Appl. 2020, 32, 17043–17057. [Google Scholar] [CrossRef]
- Quadrana, M.; Cremonesi, P.; Jannach, D. Sequence-aware Recommender Systems. In Proceedings of the 26th Conference on User Modeling, Adaptation and Personalization, Singapore, 8–11 July 2018; pp. 373–374. [Google Scholar] [CrossRef]
- Chen, D.; Hu, W.; Yuan, B.; Zhang, R.; Wang, J. Next-Item Recommendation with Deep Adaptable Co-Embedding Neural Networks. IEEE Signal Process. Lett. 2021, 28, 1220–1224. [Google Scholar] [CrossRef]
- Li, T.; Choi, M.; Fu, K.; Lin, L. Music Sequence Prediction with Mixture Hidden Markov Models. In Proceedings of the 2019 IEEE International Conference on Big Data, Los Angeles, CA, USA, 9–12 December 2019; pp. 6128–6132. [Google Scholar] [CrossRef]
- Mlika, F.; Karoui, W. Proposed Model to Intelligent Recommendation System based on Markov Chains and Grouping of Genres. Procedia Comput. Sci. 2020, 176, 868–877. [Google Scholar] [CrossRef]
- Zhang, Y.; Shi, Z.; Zuo, W.; Yue, L.; Liang, S.; Li, X. Joint Personalized Markov Chains with social network embedding for cold-start recommendation. Neurocomputing 2019, 386, 208–220. [Google Scholar] [CrossRef]
- Chen, L.; Xia, M. A context-aware recommendation approach based on feature selection. Appl. Intell. 2020, 51, 865–875. [Google Scholar] [CrossRef]
- Zhu, J.; Ma, X.; Yue, C.; Wang, C. Interest-Forgetting Markov Model for Next-Basket Recommendation. Commun. Comput. Inf. Sci. 2019, 1058, 20–31. [Google Scholar] [CrossRef]
- Aghdam, M.H. Context-aware recommender systems using hierarchical hidden Markov model. Phys. A Stat. Mech. Appl. 2018, 518, 89–98. [Google Scholar] [CrossRef]
- Wang, P.; Zhang, Y.; Niu, S.; Guo, J. Modeling Temporal Dynamics of Users’ Purchase Behaviors for Next Basket Prediction. J. Comput. Sci. Technol. 2019, 34, 1230–1240. [Google Scholar] [CrossRef]
- Nasir, M.; Ezeife, C.I.; Gidado, A. Improving e-commerce product recommendation using semantic context and sequential historical purchases. Soc. Netw. Anal. Min. 2021, 11, 82. [Google Scholar] [CrossRef]
- Wang, W.; Cao, L. Interactive Sequential Basket Recommendation by Learning Basket Couplings and Positive/Negative Feedback. ACM Trans. Inf. Syst. 2021, 39, 1–26. [Google Scholar] [CrossRef]
- Yağcı, A.M.; Aytekin, T.; Gürgen, F.S. Parallel pairwise learning to rank for collaborative filtering. Concurr. Comput. Pract. Exp. 2018, 31, e5141. [Google Scholar] [CrossRef]
- De Maio, C.; Fenza, G.; Gallo, M.; Loia, V.; Parente, M. Social media marketing through time-aware collaborative filtering. Concurr. Comput. Pract. Exp. 2017, 30, e4098. [Google Scholar] [CrossRef]
- Sharma, R.; Gopalani, D.; Meena, Y. Collaborative filtering-based recommender system: Approaches and research challenges. In Proceedings of the 2017 3rd International Conference on Computational Intelligence & Communication Technology (CICT), Ghaziabad, India, 9–10 February 2017; pp. 1–6. [Google Scholar] [CrossRef]
- Zheng, L.; Noroozi, V.; Yu, P.S. Joint Deep Modeling of Users and Items Using Reviews for Recommendation. In Proceedings of the Tenth ACM International Conference on Web Search and Data Mining, Cambridge, UK, 6–10 February 2017. [Google Scholar] [CrossRef]
- Sharma, S.; Mahajan, A. Suggestive Approaches to Create a Recommender System for GitHub. Int. J. Inf. Technol. Comput. Sci. 2017, 9, 48–55. [Google Scholar] [CrossRef]
- Raja, D.K.; Pushpa, S.; Raja, K. Novelty-driven recommendation by using integrated matrix factorization and temporal-aware clustering optimization. Int. J. Commun. Syst. 2018, 33, e3851. [Google Scholar] [CrossRef]
- Issues in various recommender system in E-commerce—A survey. J. Crit. Rev. 2020, 7, 604–608. [CrossRef]
- Çano, E.; Morisio, M. Hybrid recommender systems: A systematic literature review. Intell. Data Anal. 2017, 21, 1487–1524. [Google Scholar] [CrossRef]
- Kunaver, M.; Požrl, T. Diversity in recommender systems—A survey. Knowl.-Based Syst. 2017, 123, 154–162. [Google Scholar] [CrossRef]
- Hiriyannaiah, S.; Siddesh, G.M.; Srinivasa, K.G. DeepLSGR: Neural collaborative filtering for recommendation systems in smart community. Multimed. Tools Appl. 2022, 1–20. [Google Scholar] [CrossRef]
- Soliman, K.; Mahmood, M.A.; El Azab, A.; Hefny, H. A Survey of Recommender Systems and Geographical Recommendation Techniques. In GIS Applications in the Tourism and Hospitality Industry; IGI Global: Hershey, PA, USA, 2018; pp. 249–274. [Google Scholar] [CrossRef]
- Wang, C.-D.; Deng, Z.-H.; Lai, J.-H.; Yu, P.S. Serendipitous Recommendation in E-Commerce Using Innovator-Based Collaborative Filtering. IEEE Trans. Cybern. 2018, 49, 2678–2692. [Google Scholar] [CrossRef] [PubMed]
- Satheesan, P.; Haddela, P.S.; Alosius, J. Product Recommendation System for Supermarket. In Proceedings of the 2020 19th IEEE International Conference on Machine Learning and Applications (ICMLA), Miami, FL, USA, 14–17 December 2020; pp. 930–935. [Google Scholar] [CrossRef]
- Usmani, Z.A.; Manchekar, S.; Malim, T.; Mir, A. A predictive approach for improving the sales of products in e-commerce. In Proceedings of the 3rd IEEE International Conference on Advances in Electrical and Electronics, Information, Communication and Bio-Informatics, AEEICB 2017, Chennai, India, 27–28 February 2017; pp. 188–192. [Google Scholar] [CrossRef]
- Jia, R.; Li, R.; Yu, M.; Wang, S. E-commerce purchase prediction approach by user behavior data. In Proceedings of the IEEE CITS 2017—2017 International Conference on Computer, Information and Telecommunication Systems, Dalian, China, 21–23 July 2017; pp. 1–5. [Google Scholar] [CrossRef]
- Kommineni, M.; Alekhya, P.; Vyshnavi, T.M.; Aparna, V.; Swetha, K.; Mounika, V. Machine Learning based Efficient Recommendation System for Book Selection using User based Collaborative Filtering Algorithm. In Proceedings of the 4th International Conference on Inventive Systems and Control, ICISC 2020, Coimbatore, India, 8–10 January 2020; pp. 66–71. [Google Scholar] [CrossRef]
- Biswas, A.; Vineeth, K.S.; Jain, A. Mohana Development of Product Recommendation Engine By Collaborative Filtering and Association Rule Mining Using Machine Learning Algorithms. In Proceedings of the 4th International Conference on Inventive Systems and Control, ICISC 2020, Coimbatore, India, 8–10 January 2020; pp. 272–277. [Google Scholar] [CrossRef]
- Verma, P.; Sharma, S. Artificial Intelligence based Recommendation System. In Proceedings of the IEEE 2020 2nd International Conference on Advances in Computing, Communication Control and Networking, ICACCCN 2020, Greater Noida, India, 18–19 December 2020; pp. 669–673. [Google Scholar] [CrossRef]
- Doan, T.-N.; Sahebi, S. TransCrossCF: Transition-based Cross-Domain Collaborative Filtering. In Proceedings of the 19th IEEE International Conference on Machine Learning and Applications, ICMLA 2020, Miami, FL, USA, 14–17 December 2020; pp. 320–327. [Google Scholar] [CrossRef]
- Li, C.; Ma, L. Item-based Collaborative Filtering Algorithm Based on Group Weighted Rating. In Proceedings of the 2020 13th International Symposium on Computational Intelligence and Design, ISCID 2020, Hangzhou, China, 12–13 December 2020; pp. 114–117. [Google Scholar] [CrossRef]
- Xiao, S.; Ling, H.; Lu, Y.; Tang, Z. Study on collaborative filtering recommendation algorithm based on prediction for item rating. In Proceedings of the 2020 International Conference on Information Science, Parallel and Distributed Systems, ISPDS 2020, Xi’an, China, 14–16 August 2020; pp. 352–355. [Google Scholar] [CrossRef]
- Anwaar, F.; Iltaf, N.; Afzal, H.; Nawaz, R. HRS-CE: A hybrid framework to integrate content embeddings in recommender systems for cold start items. J. Comput. Sci. 2018, 29, 9–18. [Google Scholar] [CrossRef]
- Hwangbo, H.; Kim, Y.S.; Cha, K.J. Recommendation system development for fashion retail e-commerce. Electron. Commer. Res. Appl. 2018, 28, 94–101. [Google Scholar] [CrossRef]
- Tewari, A.S. Generating Items Recommendations by Fusing Content and User-Item based Collaborative Filtering. Procedia Comput. Sci. 2020, 167, 1934–1940. [Google Scholar] [CrossRef]
- Tang, Y.; Guo, K.; Zhang, R.; Xu, T.; Ma, J.; Chi, T. ICFR: An effective incremental collaborative filtering based recommendation architecture for personalized websites. World Wide Web 2019, 23, 1319–1340. [Google Scholar] [CrossRef]
- Jiang, M.; Zhang, Z.; Jiang, J.; Wang, Q.; Pei, Z. A collaborative filtering recommendation algorithm based on information theory and bi-clustering. Neural Comput. Appl. 2019, 31, 8279–8287. [Google Scholar] [CrossRef]
- Khoshahval, S.; Farnaghi, M.; Taleai, M.; Mansourian, A. A Personalized Location-Based and Serendipity-Oriented Point of Interest Recommender Assistant Based on Behavioral Patterns. In Lecture Notes in Geoinformation and Cartography; Springer: Berlin/Heidelberg, Germany, 2018; pp. 271–289. [Google Scholar] [CrossRef]
- Yu, H.; Wang, Y.; Fan, Y.; Meng, S.; Huang, R. Accuracy Is Not Enough: Serendipity Should Be Considered More. In Innovative Mobile and Internet Services in Ubiquitous Computing; Springer: Berlin/Heidelberg, Germany, 2017; pp. 231–241. [Google Scholar] [CrossRef]
- Gaikwad, R.S.; Udmale, S.S.; Sambhe, V.K. E-commerce Recommendation System Using Improved Probabilistic Model. In Information and Communication Technology for Sustainable Development; Springer: Singapore, 2017; pp. 277–284. [Google Scholar] [CrossRef]
- Xiao, Y.; Ezeife, C.I. E-Commerce Product Recommendation Using Historical Purchases and Clickstream Data. In Big Data Analytics and Knowledge Discovery; Springer: Berlin/Heidelberg, Germany, 2018; pp. 70–82. [Google Scholar] [CrossRef]
- Bhatta, R.; Ezeife, C.I.; Butt, M.N. Mining Sequential Patterns of Historical Purchases for E-commerce Recommendation. In Big Data Analytics and Knowledge Discovery; Springer: Berlin/Heidelberg, Germany, 2019; pp. 57–72. [Google Scholar] [CrossRef]
- He, X.; Liao, L.; Zhang, H.; Nie, L.; Hu, X.; Chua, T.-S. Neural collaborative filtering. In Proceedings of the 26th International Conference on World Wide Web, Perth, Australia, 3–7 April 2017; pp. 173–182. [Google Scholar]
- Zhu, T.; Liu, G.; Chen, G. Social Collaborative Mutual Learning for Item Recommendation. ACM Trans. Knowl. Discov. Data 2020, 14, 1–19. [Google Scholar] [CrossRef]
- Kuo, R.; Chen, C.-K.; Keng, S.-H. Application of hybrid metaheuristic with perturbation-based K-nearest neighbors algorithm and densest imputation to collaborative filtering in recommender systems. Inf. Sci. 2021, 575, 90–115. [Google Scholar] [CrossRef]
- Hu, H.; He, X.; Gao, J.; Zhang, Z.-L. Modeling Personalized Item Frequency Information for Next-basket Recommendation. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval, Virtual Event, 11–15 July 2020; pp. 1071–1080. [Google Scholar] [CrossRef]
- Gupta, G.; Katarya, R. A Study of Deep Reinforcement Learning Based Recommender Systems. In Proceedings of the 2021 2nd International Conference on Secure Cyber Computing and Communications (ICSCCC), Jalandhar, India, 21–23 May 2021; pp. 218–220. [Google Scholar] [CrossRef]
- Deldjoo, Y.; Di Noia, T.; Malitesta, D.; Merra, F.A. A Study on the Relative Importance of Convolutional Neural Networks in Visually-Aware Recommender Systems. In Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Nashville, TN, USA, 19–25 June 2021; pp. 3956–3962. [Google Scholar] [CrossRef]
- Haihan, L.; Guanglei, Q.; Nana, H.; Xinri, D. Shopping Recommendation System Design Based On Deep Learning. In Proceedings of the 2021 IEEE 6th International Conference on Intelligent Computing and Signal Processing, ICSP 2021, Xi’an, China, 9–11 April 2021; pp. 998–1001. [Google Scholar] [CrossRef]
- Khan, Z.; Hussain, M.I.; Iltaf, N.; Kim, J.; Jeon, M. Contextual recommender system for E-commerce applications. Appl. Soft Comput. 2021, 109, 107552. [Google Scholar] [CrossRef]
- Kavitha, K.; Kumar, S.L.; Pravalika, P.; Sruthi, K.; Lalitha, R.; Rao, N.K. Fashion compatibility using convolutional neural networks. Mater. Today Proc. 2020. [Google Scholar] [CrossRef]
- Reyes, L.J.P.; Oviedo, N.B.; Camacho, E.C.; Calderon, J.M. Adaptable Recommendation System for Outfit Selection with Deep Learning Approach. IFAC-PapersOnLine 2021, 54, 605–610. [Google Scholar] [CrossRef]
- Yuan, F.; Karatzoglou, A.; Arapakis, I.; Jose, J.M.; He, X. A Simple Convolutional Generative Network for Next Item Recommendation. In Proceedings of the 12th ACM International Conference on Web Search and Data Mining, Melbourne, Australia, 11–15 February 2019; pp. 582–590. [Google Scholar] [CrossRef]
- Hidasi, B.; Karatzoglou, A.; Baltrunas, L.; Tikk, D. Session-based Recommendations with Recurrent Neu-ral Networks. arXiv 2015, arXiv:1511.06939. [Google Scholar]
- Choe, B.; Kang, T.; Jung, K. Recommendation System with Hierarchical Recurrent Neural Network for Long-Term Time Series. IEEE Access 2021, 9, 72033–72039. [Google Scholar] [CrossRef]
- Rabiu, I.; Salim, N.; Da’U, A.; Nasser, M. Modeling sentimental bias and temporal dynamics for adaptive deep recommendation system. Expert Syst. Appl. 2021, 191, 116262. [Google Scholar] [CrossRef]
- Wang, P.; Li, Z.; Zhang, Y.; Hou, Y.; Ge, L. QPIN: A Quantum-inspired Preference Interactive Network for E-commerce Recommendation. In Proceedings of the International Conference on Information and Knowledge Management, Beijing, China, 3–7 November 2019; pp. 2329–2332. [Google Scholar] [CrossRef]
- Lo, C.; Yu, H.; Yin, X.; Shetty, K.; He, C.; Hu, K.; Platz, J.M.; Ilardi, A.; Madhvanath, S. Page-level Optimization of e-Commerce Item Recommendations. In Proceedings of the RecSys 2021—15th ACM Conference on Recommender Systems, Amsterdam, The Netherlands, 27 September–1 October 2021; pp. 495–504. [Google Scholar] [CrossRef]
- Han, J.; Ma, Y.; Mei, Q.; Liu, X. DeepRec: On-device Deep Learning for Privacy-Preserving Sequential Recommendation in Mobile Commerce. In Proceedings of the World Wide Web Conference, WWW 2021, Ljubljana, Slovenia, 19–23 April 2021; pp. 900–911. [Google Scholar] [CrossRef]
- Louis, M.; Azad, Z.; Delhadtehrani, L.; Gupta, S.L.; Warden, P.; Reddi, V.; Joshi, A. Towards Deep Learning using TensorFlow Lite on RISC-V. In Proceedings of the Third Workshop on Computer Architecture Research with RISC-V (CARRV 2019), Phoenix, AZ, USA, 22 June 2019. [Google Scholar] [CrossRef]
- Liu, W.; Zhang, Y.; Wang, J.; He, Y.; Caverlee, J.; Chan, P.P.K.; Yeung, D.S.; Heng, P.-A. Item Relationship Graph Neural Networks for E-Commerce. IEEE Trans. Neural Netw. Learn. Syst. 2021, 1–15. [Google Scholar] [CrossRef]
- Wong, C.-M.; Feng, F.; Zhang, W.; Chen, H.; Zhang, Y.; He, P.; Chen, H.; Zhao, K. Improving Conversational Recommender System by Pretraining Billion-scale Knowledge Graph. In Proceedings of the 2021 IEEE 37th International Conference on Data Engineering (ICDE), Chania, Greece, 19–22 April 2021; pp. 2607–2612. [Google Scholar] [CrossRef]
- Tao, Y.; Wang, C.; Yao, L.; Li, W.; Yu, Y. Item trend learning for sequential recommendation system using gated graph neural network. Neural Comput. Appl. 2021, 1–16. [Google Scholar] [CrossRef]
- Zhang, Y.; Yuan, M.; Zhao, C.; Chen, M.; Liu, X. Integrating label propagation with graph convolutional networks for recommendation. Neural Comput. Appl. 2022, 34, 8211–8225. [Google Scholar] [CrossRef]
- Zhao, K.; Zheng, Y.; Zhuang, T.; Li, X.; Zeng, X. Joint Learning of E-commerce Search and Recommendation with a Unified Graph Neural Network. In Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining, Melbourne, Australia, 11–15 February 2022; pp. 1461–1469. [Google Scholar] [CrossRef]
- Wang, H.; Amagata, D.; Makeawa, T.; Hara, T.; Hao, N.; Yonekawa, K.; Kurokawa, M. A DNN-Based Cross-Domain Recommender System for Alleviating Cold-Start Problem in E-Commerce. IEEE Open J. Ind. Electron. Soc. 2020, 1, 194–206. [Google Scholar] [CrossRef]
- Ahmed, A.; Saleem, K.; Khalid, O.; Rashid, U. On deep neural network for trust aware cross domain recommendations in E-commerce. Expert Syst. Appl. 2021, 174, 114757. [Google Scholar] [CrossRef]
- Abinaya, S.; Devi, M.K.K. Enhancing Top-N Recommendation Using Stacked Autoencoder in Context-Aware Recommender System. Neural Process. Lett. 2021, 53, 1865–1888. [Google Scholar] [CrossRef]
- Qin, Y.; Wang, P.; Li, C. The World is Binary: Contrastive Learning for Denoising Next Basket Recommendation. In Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, Montreal, QC, Canada, 12–15 July 2021; pp. 859–868. [Google Scholar] [CrossRef]
- Ngaffo, A.N.; Choukair, Z. A deep neural network-based collaborative filtering using a matrix factorization with a twofold regularization. Neural Comput. Appl. 2022, 34, 6991–7003. [Google Scholar] [CrossRef]
- Liu, T.; He, Z. DLIR: A deep learning-based initialization recommendation algorithm for trust-aware recommendation. Appl. Intell. 2022, 1–12. [Google Scholar] [CrossRef]
- Chorowski, J.; Bahdanau, D.; Serdyuk, D.; Cho, K.; Bengio, Y. Attention-Based Models for Speech Recognition. In Proceedings of the 29th Annual Conference on Neural Information Processing Systems 2015, Montreal, QC, Canada, 7–12 December 2015. [Google Scholar]
- Zhang, J.; Wang, D.; Yu, D. TLSAN: Time-aware long- and short-term attention network for next-item recommendation. Neurocomputing 2021, 441, 179–191. [Google Scholar] [CrossRef]
- Thaipisutikul, T.; Shih, T.K. A novel context-aware recommender system based on a deep sequential learning approach (CReS). Neural Comput. Appl. 2021, 33, 11067–11090. [Google Scholar] [CrossRef]
- Che, B.; Zhao, P.; Fang, J.; Zhao, L.; Sheng, V.S.; Cui, Z.; Sheng, V. Inter-Basket and Intra-Basket Adaptive Attention Network for Next Basket Recommendation. IEEE Access 2019, 7, 80644–80650. [Google Scholar] [CrossRef]
- Liu, T.; Yin, X.; Ni, W. Next Basket Recommendation Model Based on Attribute-Aware Multi-Level Attention. IEEE Access 2020, 8, 153872–153880. [Google Scholar] [CrossRef]
- Da’U, A.; Salim, N. Sentiment-Aware Deep Recommender System With Neural Attention Networks. IEEE Access 2019, 7, 45472–45484. [Google Scholar] [CrossRef]
- Ouyang, S.; Lawlor, A. Improving Explainable Recommendations by Deep Review-Based Explanations. IEEE Access 2021, 9, 67444–67455. [Google Scholar] [CrossRef]
- Cui, Q.; Wu, S.; Huang, Y.; Wang, L. A hierarchical contextual attention-based network for sequential recommendation. Neurocomputing 2019, 358, 141–149. [Google Scholar] [CrossRef]
- Zhang, L.; Wang, P.; Li, J.; Xiao, Z.; Shi, H. Attentive Hybrid Recurrent Neural Networks for sequential recommendation. Neural Comput. Appl. 2021, 33, 11091–11105. [Google Scholar] [CrossRef]
- Li, W.; Xu, B. Aspect-Based Fashion Recommendation With Attention Mechanism. IEEE Access 2020, 8, 141814–141823. [Google Scholar] [CrossRef]
- Du, Y.; Wang, L.; Peng, Z.; Guo, W. Review-based hierarchical attention cooperative neural networks for recommendation. Neurocomputing 2021, 447, 38–47. [Google Scholar] [CrossRef]
- Zhu, H.; Li, X.; Zhang, P.; Li, G.; He, J.; Li, H.; Gai, K. Learning Tree-based Deep Model for Recommender Systems. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, London, UK, 19–23 August 2018; pp. 1079–1088. [Google Scholar] [CrossRef]
- Harper, F.M.; Konstan, J.A. The MovieLens Datasets: History and context. ACM Trans. Interact. Intell. Syst. 2015, 5, 1–19. [Google Scholar] [CrossRef]
- Singer, U.; Roitman, H.; Eshel, Y.; Nus, A.; Guy, I.; Levi, O.; Hasson, I.; Kiperwasser, E. Sequential Modeling with Multiple Attributes for Watchlist Recommendation in E-Commerce. In Proceedings of the 15th ACM International Conference on Web Search and Data Mining, Virtual, 21–25 February 2022; pp. 937–946. [Google Scholar] [CrossRef]
- Chai, T.; Draxler, R.R. Root mean square error (RMSE) or mean absolute error (MAE)?—Arguments against avoiding RMSE in the literature. Geosci. Model Dev. 2014, 7, 1247–1250. [Google Scholar] [CrossRef]
- Ni, J.; Li, J.; McAuley, J. Justifying Recommendations using Distantly-Labeled Reviews and Fine-Grained Aspects. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Hong Kong, China, 3–7 November 2019; pp. 188–197. [Google Scholar]
- Zhang, M.-L.; Zhou, Z.-H. A review on multi-label learning algorithms. IEEE Trans. Knowl. Data Eng. 2013, 26, 1819–1837. [Google Scholar] [CrossRef]
- Jannach, D.; Lerche, L.; Jugovac, M. Adaptation and Evaluation of Recommendations for Short-term Shopping Goals. In Proceedings of the 9th ACM Conference on Recommender Systems, Vienna, Austria, 16–20 September 2015; pp. 211–218. [Google Scholar] [CrossRef]
- Castells, P.; Vargas, S.; Wang, J. Novelty and Diversity Metrics for Recommender Systems: Choice, Discovery and Relevance; Universidad Autónoma de Madrid: Madrid, Spain, 2011. [Google Scholar]
- Breese, J.S.; Heckerman, D.; Kadie, C. Empirical Analysis of Predictive Algorithms for Collaborative Fil-tering. arXiv 2013, arXiv:1301.7363. [Google Scholar]
Study | Focus | Publication Year | Studies Reviewed | Years Covered |
---|---|---|---|---|
Alyari and Jafari [12] | Recommender systems | 2018 | 51 | 2005–2018 |
Portugal et al. [13] | Recommender systems | 2018 | 121 | 2001–2016 |
Monti et al. [14] | Multicriteria recommender systems | 2021 | 93 | 2003–2018 |
Villegas et al. [15] | Context-aware recommenders | 2017 | 87 | 2004–2016 |
Murciego et al. [16] | Context-aware recommender systems in the music domain | 2021 | 100 | 2010–2021 |
Jesse and Jannach [17] | Digital nudging in recommender systems | 2021 | 63 | 2009–2021 |
Da’u and Salim [18] | Deep learning methods in recommender systems | 2019 | 99 | 2007–2018 |
Khan et al. [19] | Recommender systems for e-tourism | 2021 | 143 | 2012–2020 |
Hamid et al. [20] | Recommender systems for e-tourism | 2021 | 65 | 2013–2020 |
Mohammadi et al. [21] | Trust-based recommender systems in Internet of Things | 2019 | 59 | 2011–2018 |
Rahayu et al. [22] | Use of ontologies in recommender systems | 2022 | 28 | 2010–2020 |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[31] | Considering interactions in the historical sessions and the changing semantics of an item over time | MC | Hybrid representations allow dynamic attention to each factor while maintaining a balance between general preferences and sequential patterns | Sparsity | IEEE/J/2021 |
[32] | Comparing the mixed model with state-of-the-art methods for evaluating the music predictions | Hidden Markov model (HMM) | Developing a large-scale real-world dataset in a Kaggle competition | Poor Markov property for general cases; Missing real-time user feedback | IEEE/CP/2019 |
[33] | Proposing an intelligent recommender system | MC and grouping genres | The merger of two different techniques improves the quality of RS | Data sparsity | SD/J/2020 |
[34] | Utilizing social and temporal information in RS | MC | Addresses user cold-start issue by a two-level model founded on MC at both user-level and user group level to consider user picks dynamically | Data sparsity | SD/J/2019 |
[35] | Developing a context-aware recommendation approach to handle data sparsity issue | MC/CARS | Embedded features selection to handle the data sparsity issue | Scalability | Springer/J/2021 |
[36] | Solving the data sparsity problem of the transition matrix | MC | Integrating interest-forgetting attributes, social trust relation and item similarity in personalized Markov model | Single-user transition matrix data sparseness; cold start | Springer/C/2019 |
[37] | Exploiting the hierarchical hidden Markov model (HHMM) to extract latent context from the data | Hidden Markov model | The proposed model significantly improves performance both in terms of accuracy and the diversity of recommendations | Scalability | Springer/J/2019 |
[38] | Proposing sequential offers to users based on a succession of the user’s prior reaction to recommendations | MC | The study proposes the dynamic model learning technique for a time-dependent next-basket recommendation, which together sports the dynamic character of user selections and item relations | Sparsity; scalability | Springer/J/2019 |
[39] | Handling personalization and a lack of semantics among recommended items | MC | The study proposes a model to enhance e-commerce product recommendation using semantic context and sequential chronological acquisitions | Scalability | Springer/J/2021 |
[40] | Interactive sequential basket recommendation, which iteratively predicts next baskets by learning the intrabasket/interbasket couplings between items and both positive and negative user feedback on recommended baskets | MC | Intrabasket/interbasket couplings, incorporating the user selection and nonselection | A large number of items in the next basket lead to unfair comparison, underfitting, and poor performance | ACM/J/2021 |
Techniques | Representative Algorithm | Advantages | Disadvantages |
---|---|---|---|
Memory-Based Collaborative Filtering (CF) | User-Based CF; Item-Based CF | Efficient scalability; Easy to implement; Adding data is straightforward; Content is not a factor | Cold start, data sparsity and scalability issues; Relying on the precise recommendations; Biased for massive dataset |
Model-Based Collaborative Filtering | Slope one CF | Improves the prediction efficiency; Enhances sparsity and scalability issues | Model is pricey; Loss of details in a factorization matrix |
Hybrid Collaborative Filtering | A blend of memory-based and model-based | Controls the sparsity constraints; Enhances the prediction efficiency | Raised complexity; Challenging for implementation |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[52] | Proposed innovator-based collaborative filtering (INVBCF) for recommending cold items | Collaborative filtering | Cold items can be managed in the recommendation list via innovators, achieving the balance between serendipity and accuracy | Poor performance on sparse data | IEEE/J/2019 |
[53] | To help customers acquire recommendations on the purchase of desired products | Collaborative filtering, clustering, and association rule mining | The item-to-item collaborative filtering is selected as it furnishes recommendations to all customers regardless of the number of buyers and in the ratings given already on priority | Cold start, data sparsity | IEEE/CP/2020 |
[54] | This article concerns diverse issues faced by recommender systems and proposes solutions to them | Classification, collaborative filtering, association rule mining and sequence rule mining | User classification | Cold start, data sparsity, scalability | IEEE/CP/2017 |
[55] | Users’ purchase prediction using clicking behavior features | CF | Evaluation of the proposed system with the dataset provided by Ali Mobile Recommendation Competition held in 2015 | Cold start | IEEE/CP/2017 |
[56] | A book recommender system | User-based CF | A user-user similarity matrix | Cold start, data sparsity | IEEE/CP/2020 |
[57] | To make accurate and efficient recommendations | CF | Multiple algorithms are used to generate user-based recommendations and item-based recommendations | Cold start, data sparsity | IEEE/CP/2020 |
[58] | AI-based recommender system | CF | A thorough review of CF-based recommender systems | Data sparsity | IEEE/CP/2020 |
[59] | Transition-based cross-domain collaborative filtering | CF | Overcomes the cold-start problem | Cross-domain review analysis, data sparsity | IEEE/CP/2020 |
[60] | Improvement of item-based CF algorithm | CF | Proposed group weighted rating method to improve item-based CF | Cold start, data sparsity, scalability | IEEE/CP/2020 |
[61] | Enhancement of item-based CF recommendation | CF | Proposed similarity-based algorithm using modified cosine similarity algorithm | Cold start | IEEE/CP/2020 |
[62] | Predicting the ratings for cold-start items by exploiting items’ textual descriptions | CF | The study focuses on the item cold-start problem | User cold-start problem, data sparsity, scalability | SD/J/2018 |
[63] | Suggesting fashion goods for clients by enhancing the existing CF procedure to consider the features of fashion products | CF | Both online and offline purchase data used to generate recommendations | Experimental results cannot demonstrate the effects of using online and offline data, as well as the effects of a decline in preference over time | SD/J/2018 |
[64] | Curtailing the size of the recommendation list | CF | Used item weight generator block to rank recommended items at the appropriate position in the list | Cold start, data sparsity | SD/J/2020 |
[65] | Considering users’ retrieval intentions for more personalized recommendations | CF | Incremental strategy for the collaborative filtering recommendation algorithm | Cold start, data sparsity | Springer/J/2020 |
[66] | To overcome the data sparsity problem of the CF algorithm | CF | Optimizing biclustering and information entropy | Scalability, cold start | Springer/J/2019 |
[67] | Serendipity-oriented location-based recommender system | Association rule mining and CF | A personalized recommender assistant which offers both precise and spontaneous points of interest (POIs) | Cold start | Springer/CP/2018 |
[68] | Striking a balance between accuracy and serendipity | CF | The proposed strategy employed user records and feedback to optimize and balance serendipity and accuracy | Cold start | Springer/CP/2017 |
[69] | Improving search query analysis by using collaborative filtering and naive Bayes algorithms | CF, naïve Bayes | Improved F1 measure by 14% compared to simple naïve Bayes technique | Cold start, data sparsity | Springer/CP/2017 |
[70] | Improving the quality of the user–item matrix by normalizing the frequency of item purchase | Clickstream-based CF recommender system | Improved quality of ratings | Poor integration between sequential patterns of the user–item matrix. | Springer/J/2018 |
[71] | Developing a hybrid recommender system | CF | Improved recommendations through better customer behavior modeling and enhanced user–item matrix | Inability to incorporate multiple data source-based sequential patterns; no provision for infrequent users | Springer/J/2019 |
[72] | Proposing a neural network-based collaborative filtering framework | CF | The proposed model merges the strength of the linearity of MF and the nonlinearity of multilayer perceptron for modeling the user–item latent structures | Data sparsity | ACM/J/2017 |
[73] | Implementing social, collaborative mutual learning for recommender system | CF | Combining item-based CF and social CF to improve the recommendation performance | Scalability | ACM/J/2020 |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[74] | Addressing the problems of data sparsity and similarities selection | kNN with the densest imputation | Model simplicity | High computational time to build the model; Low scalability; Limited validation | SD/J/2021 |
[75] | Enhancing NBR recommendation by capturing two important patterns associated with personalized item frequency (PIF) | kNN | Effective use of personalized item frequency (PIF) with a simple method | Low scalability | ACM/C/2020 |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[78] | Improving recommendations using salient visual features of products | CNN | Use of real-user and commodity dataset | Low scalability; High operation cost; Limited parameters considered | IEEE/C/2021 |
[79] | Addressing the sparsity problem | CNN, Word2Vec | Concurrent handling of semantics and contexts of user and item data | Temporal effect not addressed; Data noise issue may seriously affect performance | SD/J/2021 |
[80] | Offering a recommender system based on text and image inputs | CNN | Simplicity of the system. | Low scalability; Limited features considered; Limited validation | SD/C/2020 |
[81] | Improving recommendations using the score from past interactions | CNN | Simplicity of the system. | Low scalability; Limited features considered; Limited validation | SD/C/2020 |
[82] | To keep an ordered collection of all items the user interacts in a session | Generative CNN | Effectively addresses the problems with a typical session-based CNN recommender | Low scalability; Limited use for long sequences of user feedback | ACM/C/2019 |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[84] | Enhancing the use of temporal properties of the user history | Multilayered RNN, long- and short-term time series | Effectively uses temporal properties in time-series data to improve recommendation quality | Review content and ratings not considered; Low scalability | IEEE/J/2021 |
[31] | Balancing the combination of general preference and the sequential patterns factors in recommendations | LSTM | Deducing high-level preferences of users based on the observed sessions | High operation cost; Limited validation | IEEE/J/2021 |
[85] | Dealing with problems of data unbalance and sparsity | LSTM | Dynamic modeling of user and item features | High operation cost | SD/J/2022 |
[86] | Learning user preferences by leveraging the mathematical formalisms of quantum theory | RNN, quantum many-body wave function (QMWF) | The mathematical formalism of QT is exploited | Low scalability; Limited validation | ACM/C/2019 |
[87] | Enhancing recommendations for relevant items on the items details page and increasing discoverability of relevant items | Two-stage RNN | Formulates page-level optimization problem as a sequential ranking problem and accommodates heterogeneous feature types | Low scalability; Limited validation | ACM/C/2021 |
[88] | Enhancing sequential recommendation while ensuring privacy by sending no raw data out of the mobile device | RNN, Automated Gradual Pruner (AGP) | Addresses privacy as the primary concern and addresses issues specific to mobile devices | Low scalability; Limited validation | ACM/C/2021 |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[90] | Improving product relationship representation by exploiting the graph’s topological structures | Graph neural network, edge relational network, multihop dependencies | Improves inference of complex dependencies in item relationships | Limited validation | IEEE/J/2021 |
[91] | Solving the problems of data scarcity and sparseness related to click-through rate prediction | Conversation knowledge graph, deep convolution network | Utilizes various types of information from user states and dialogue interactions | Low scalability; Entity descriptions and knowledge schema are not considered | IEEE/C/2021 |
[92] | Improving the modeling of the current item by incorporating the item trend information | Gated graph neural network, self-attention layer | Item trend information from the implicit history of the user incorporated into subsequent recommendation tasks | Low scalability; Cold-start problem | Springer/J/2021 |
[93] | Solving the problem of over-smoothing of graph convolutional networks to improve the recommendation performance | Graph convolutional network, label propagation algorithm, attention network | Improved recommendation performance by the unification of GCN and LPA | Static preference modeling; Low training performance | Springer/J/2022 |
[94] | Alleviate the sparseness of user behaviors by jointly learning from search and recommendation scenarios | Graph neural network with the aggregation layer | Search and recommendation scenarios are jointly learned | Low scalability; Limited validation | ACM/C/2022 |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[95] | Alleviating the cold-start problem | DNN, Stacking model, Word2Vec | Enhances user modeling | High operation cost; Limited validation | IEEE/J/2020 |
[96] | Alleviating the cold-start problem | DNN, latent factor model, multilayer perceptron | Models relationships in cross-domain scenarios; Generates useful recommendations in high sparsity scenarios | Implementation complexity | SD/J/2021 |
[97] | Deriving sentiment by merging ratings and reviews to enhance top-n recommendations | Context-specific sentiment-based stacked encoder (CSSAE), item splitting | Computing concrete preferences of the user by integrating rating and review for a given context | Low scalability | Springer/J/2021 |
[99] | Solving problems of data sparseness and cold start to increase recommendation quality | Matrix factorization, combination of multilayer and single-layer perceptron | Effectively solves issues with latent factors | Low scalability; Limited validation | Springer/J/2022 |
[100] | Solving the problem of random initialization of latent features, which play an important role in obtaining a good local minimum | Deep autoencoder, social-trust holistic learning model, item classification | Effective integration of users and items in latent feature vectors | High operation cost; Limited validation | Springer/J/2022 |
[98] | Enhancing NBR by denoising the baskets and extracting credibly relevant items | Context encoder, contrastive learning | Item-level denoising for a basket | Low scalability; Limited validation | ACM/C/2021 |
Study | Objective | Method/Model/Features | Strength | Weakness | Source |
---|---|---|---|---|---|
[102] | Utilizing knowledge from aggregations of users’ behavior and taste records to deal with sparsity and improve recommendations | Attention layers with position embeddings | Effective time-sensitive next-item recommendation; Low model complexity with high parallelism. | Low scalability | SD/J/2021 |
[103] | Improving the extraction of sequential behaviors from contextual info and representation of users’ short-term preferences | Hierarchical attention network, neural attentive bidirectional GRU with sequential residual connection | Gradual refining of user’s preferences via a hierarchical learning process | High operation cost; Limited context information | Springer/J/2020 |
[104] | Addressing the problems resulting from static use of attention-based RNNs | Adaptive attention-based RNN | Acts on all historical baskets as well as at item level in the most recent basket | High operation cost; Low scalability | IEEE/J/2019 |
[106] | Enhancing the use of user sentiments in reviews to improve the performance of recommendation | Attention-based RNN, neural coattention | Incorporation of sentiments with consideration of various aspects of a product | High operation cost | IEEE/J/2019 |
[105] | Incorporating the context (short-time interests and long-term preferences) to improve recommendations | Multilevel attention, item attributes | Enhanced basket-level recommendation performance | Low scalability; Minimal attributes information | IEEE/J/2020 |
[108] | Alleviating the limitations of RNNs in the acquisition of the short-term interest | Attention-based RNN, contextual hidden states | Contextual modeling of interest-based on recent factors | Does not explicitly model long-term interest | SD/J/2019 |
[109] | Capturing the user’s general preferences as well as his/her latest intent | Hybrid network with Bi-LSTM and GRU modules | Combining the usual preference with the latest intent | High operation cost; Low training efficiency | Springer/J/2021 |
[110] | Improving the local and global aspect representations obtained from user reviews | Context embedding, local and global aspects FE | Incorporates preferences in different aspects; Determines local and global importance of each word in review | Computational expense; Synonymity problems | IEEE/J/2020 |
[107] | Generating compelling review-based explanations for input to reviewer-aware recommenders | Attention-based RNN | High prediction performance; Deals with item sparsity problems | Synonymity problems; Limited validation | IEEE/J/2021 |
[111] | Enhancing user and item feature representation | Hierarchical attention, CNN | High prediction performance on real-world datasets | High operation cost | SD/J/2021 |
Name | Used By | No. of Users (1000/s) | No. of Items (1000/s) | No. of Records (Millions) | Availability | URL | Year |
---|---|---|---|---|---|---|---|
Taobao18 | [93] | 988 | 4162 | 100 | Public | https://tianchi.aliyun.com/dataset/dataDetail?dataId=649, (accessed on 30 May 2022) | 2018 |
UserBehavior | [112] | 970 | 4158 | 100 | Public | https://tianchi.aliyun.com/dataset/dataDetail?dataId=649&userId=1, (accessed on 30 May 2022) | 2017 |
Taobao20 | [94] | 83 | 9621 | 75 | On request | taobao.com, (accessed on 30 May 2022) | 2020 |
MovieLens | [113] | 280 | 58 | 27 | Public | https://files.grouplens.org/datasets/movielens/ml-latest.zip, (accessed on 30 May 2022) | 2018 |
eBay | [114] | 40 | 5375 | 12 | Public | github.com/urielsinger/Trans2D, (accessed on 30 May 2022) | 2021 |
ValuedShopper | [98] | 10 | 8 | 5 | Public | https://www.kaggle.com/competitions/acquire-valued-shoppers-challenge/data, (accessed on 30 May 2022) | 2014 |
Instacart | [98] | 200 | 8 | 3 | Public | https://www.instacart.com/datasets/grocery-shopping-2017, (accessed on 30 May 2022) | 2017 |
Ali Express | [96] | 1507 | 49 | 2 | On request | https://doi.org/10.1016/j.eswa.2021.114757, (accessed on 30 May 2022) | 2021 |
Yelp | [106] | 169 | 63 | 2 | Public | https://www.yelp.com/dataset, (accessed on 30 May 2022) | 2021 |
Outbrain | [108] | 66 | 69 | 0.83 | Public | https://www.kaggle.com/c/outbrain-click-prediction, (accessed on 30 May 2022) | 2017 |
Epinions | [100] | 49 | 140 | 0.66 | Public | trustlet.org/downloaded_epinions.html, (accessed on 30 May 2022) | 2003 |
Dunnhumby | [115] | 36 | 5 | 0.52 | Public | https://www.dunnhumby.com/source-files/, (accessed on 30 May 2022) | 2014 |
Ta-Feng | [104] | 14 | 12 | 0.50 | Public | https://www.kaggle.com/datasets/chiranjivdas09/ta-feng-grocery-dataset, (accessed on 30 May 2022) | 2001 |
Amazon | [116] | 58 | 50 | 0.32 | Public | http://jmcauley.ucsd.edu/data/amazon/, (accessed on 30 May 2022) | 2018 |
JingDong | [104] | 10 | 9 | 0.21 | Public | https://www.datafountain.cn/competitions/247/datasets, (accessed on 30 May 2022) | 2018 |
Book-crossing | [93] | 18 | 15 | 0.14 | Public | http://www2.informatik.uni-freiburg.de/~cziegler/BX/, (accessed on 30 May 2022) | 2004 |
Last.fm | [93] | 2 | 18 | 0.09 | Public | https://grouplens.org/datasets/hetrec-2011/, (accessed on 30 May 2022) | 2011 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ilyas, Q.M.; Mehmood, A.; Ahmad, A.; Ahmad, M. A Systematic Study on a Customer’s Next-Items Recommendation Techniques. Sustainability 2022, 14, 7175. https://doi.org/10.3390/su14127175
Ilyas QM, Mehmood A, Ahmad A, Ahmad M. A Systematic Study on a Customer’s Next-Items Recommendation Techniques. Sustainability. 2022; 14(12):7175. https://doi.org/10.3390/su14127175
Chicago/Turabian StyleIlyas, Qazi Mudassar, Abid Mehmood, Ashfaq Ahmad, and Muneer Ahmad. 2022. "A Systematic Study on a Customer’s Next-Items Recommendation Techniques" Sustainability 14, no. 12: 7175. https://doi.org/10.3390/su14127175